everyayah-phonemes

Phoneme-labelled Quran Datatset This dataset contains recitations from 45 professional Quran reciters, sourced from EveryAyah and QUL. The audio has been automatically phoneme-labelled using a custom phonemizer that encodes Tajweed rules. Dataset Structure audio: 16 kHz resampled mono audio duration: length of the audio in seconds verse: reference in {surah num} {ayah num} format reciter: name of the Qari’ text: diacritised text in Uthmani script phonemes:… See the full description on the dataset page:

Source: Hugging Face Hub (hetchyy/everyayah-phonemes). Metadata imported from the dataset’s Hub tags.